The sad story about the risks of power morcellation for the treatment of fibroids has received considerable media attention. The state of play through July, 2014 was described in a series of articles in the Cancer Letter of July 4, 2014. (Look here.)
Uterine fibroids are a common affliction of women. Their preferred surgical management had changed from open surgery to minimally invasive surgery, sometimes robotically performed, and often incorporating a device called a power morcellator to pulverize the fibroids, allowing the use of small incision. However, after a patient at the Brigham and Womens' Hospital (BWH) in Boston, part of the Partners Healthcare system, was found to have disseminated sarcoma post power morcellation, most likely because the machine spread tumor cells throughout her abdomen, the US Food and Drug Administration (FDA) issued an advisory against using the procedure morcellation, and despite some controversy, the procedure is now uncommonly used.
The patient so severely affected was Dr Amy Reed, an anesthesiologist at the BWH. Her husband, a cardiothoracic surgeon at BWH, launched a campaign to reduce the use of power morcellation. His pursuit of this campaign underlined an important part of the history,
The use of power morcellation was effectively allowed by the FDA in the absence of any controlled trials meant to assess its safety or efficacy in this context. The device was deemed moderate risk and approved through the 510(k) process because it appeared similar to previously allowed devices, even though older devices were not used to remove fibroids. The Cancer Letter quoted Dr David Challoner, who was on a relevant Institute of Medicine (IOM) committee, saying allowing the device on the market was
one more example of the clearance of a device for a use, not approval, based on predicates already in the market, that is, prior morcellators for other uses
The New England Journal of Medicine Weighs In - with a Special Pleading
Once again, NEJM national correspondent Dr Lisa Rosenbaum had a contrarian view(1). As we discussed here and here, last year Dr Rosenbaum wrote three commentaries suggesting the importance of conflicts of interest in health care had been overblown, especially with respect to medical journals.
This time, she argued that the FDA overreacted to the tragic case of Dr Amy Reed. In particular, she was afraid that the risks of power morcellation were exaggerated, and based on poor quality data,
Several experts argued that these risk estimates were too high and that it was riskier to expose 100,000 or so women per year to open procedures rather than laparoscopic ones. Since the rarity of LMS precludes a randomized trial, however, risk estimates had to be based primarily on retrospective case series of varying rigor. Some studies were poorly stratified for risk factors such as age, and others spanned decades during which diagnostic criteria for LMS had changed.
Dr Rosenbaum failed to mention that the lack of good data about the risks of power morcellation stemmed from the lack of any large, well designed randomized controlled trials done to assess its benefits and harms. Of course, since the device's use was allowed by the FDA 510(k) process without any requirement for trial data, there was no incentive for device companies, at least, to do such trials.
Nonetheless, Dr Rosenbaum also argued the benefits of power morcellation were downplayed.
the benefits of morcellation are largely invisible and thus 'unavailable.' Who sees the women who undergo a minimally invasive procedure, recover quickly, and avoid losing income? What does a pulmonary embolus, a wound infection, or a hemorrhage that didn’t happen look like? You can’t post pictures of these nonevents on social media. But their nonoccurrence is why we ought to be celebrating.
A March 18, 2016 article asked my opinion about Dr Rosenbaum's use of the clinical evidence.
'She talks about the data concerning the possible harms of power morcellation and argues that that data comes from relatively low-quality studies, not randomized, controlled trials, and that it is hard to tell the actual rate of harms, in particular, the dissemination of cancer. On the other hand, Dr. Rosenbaum implies that the benefits of power morcellation are well known.'
'She writes initially that power morcellation allows the treatment of fibroids to be ‘done more efficiently and effectively,’ she later implies that power morcellation is less invasive, leads to quicker recovery, avoids income loss, and furthermore, reduces the likelihood of pulmonary embolus, wound infection, or hemorrhage.'
'However, she doesn’t provide any data about these ostensible benefits. A quick search suggested that there are no good randomized, controlled trials that assessed benefits. Her argument that we have abandoned a beneficial treatment based on poor quality and perhaps exaggerated the data about its harms—that does not seem to be supported by any clear data about its benefits.'
I rendered that opinion on March 18 before I read a Cochrane Collaboration review of minimally invasive surgery versus open surgery for fibroids(2). However, that review does not seem to contradict my statements above. The review included some patients who were treated using power morcellation, but apparently did not find any studies that assessed patients treated only with power morcellation with those treated only with an alternative. Even so, it only included nine studies that enrolled a total of 808 patients. Thus, I think it is reasonable to say that there have not been any large, well-done randomized controlled trials of power morcellation versus other treatments of fibroids. Even so, while the review found some advantages for minimally invasive surgery versus open surgery in terms of short-term post-operative pain and length of hospital stay, it did not address pulmonary embolus, wound infection, or hemorrhage directly. So while Dr Rosenbaum was right to say that the evidence from clinical research about the magnitude and nature of harms of power morcellation was relatively weak, the evidence about its benefits is also weak. Thus, Dr Rosenbaum's argument that the harms have been exaggerated while the benefits were overlooked was based on a logical fallacy, special pleading.
This special pleading seemed the basis for her claim that
women may suffer more from its [power morcellation's] disuse.
So, as I was quoted by the Cancer Letter,
'The NEJM is perhaps the most prestigious, most highly regarded English-language medical journal in the world,' Poses said. 'It is, in many cases, viewed as the standard for scholarly medical journals. I am a bit surprised that it published a commentary by its own national correspondent that appears to make an argument—about benefits and harms of treatment and policymaking about treatment—that does not have a clear discussion of the data that support or fail to support either the benefits or the harms of the treatment.'
More Arguments, Less Justification
Dr Rosenbaum insisted that her concerns were about the question,
How do you use data to clarify tough trade-offs when the most compelling narratives paint evidence-based reasoning itself as an anathema?
However, she did not demonstrate an approach to policy making on power morcellation that was more evidence-based than what has transpired so far.
In addition, Dr Rosenbaum decried challenges to health care innovation from "the power of tragic stories," and the title of her commentary ("N-of-1 policymaking") suggested that the the FDA approach to power morcellation was based on a single tragic story. Yet an FDA spokesperson insisted in the March 18, 2016 Cancer Letter article,
The FDA evaluated the available data at the time and determined that it was of sufficient quality and reliability to support our November 14, 2014 decision.
Furthermore, Dr Rosenbaum expressed concerns that power morcellation met its "demise" because of unwarranted concerns about the "greedy corporation," "medicine's corruption," and "industry greed." Yet she ignored arguments that these concerns were not unreasonable.
First, the commentary ignored claims that power morcellation procedures were very lucrative, e.g., those in the 2014 Cancer Letter article
'This is a very lucrative procedure.' [Dr] Noorchashm, [Dr Reed's husband] said. 'The procedure itself bills $30,000 to $50,000, depending on the center.'
Nor did she argue against the assertion in the same article that device companies used political influence to promote their very expensive product,
There is a very strong pressure from the device industry to get into the market quickly.
The device companies have been able to make [the 510(k) process that allowed power morcellation] survive politically over the ensuing 40 years.
Also, Dr Rosenbaum's NEJM commentary failed to counter the assertion by Dr Noorchashm that the device manufacturers concealed the risks of power morcellation.
Device manufacturers clearly knew of the cancer risk. You can see warnings about malignant tumors in the Ethicon and Karl Storz user manuals. Clearly, their lawyers had warned them to put them there to avoid liability. The bottom line is that these manufacturers knew of this hazard, but neither reported it back to the FDA, as would have been the safe and responsible thing to do.(3)
Finally, according to the March 18, 2016 Cancer Letter article, while Dr Reed has apparently sued the BWH, and Dr Rosenbaum admitted she is on "faculty" at the BWH, neither she, the NEJM, nor the BWH will say whether she is currently being paid by the BWH. Particularly,
the hospital declined to provide information on Rosenbaum's title and whether she is a full-time faculty member, citing personnel policies.
My response (in the Cancer Letter article) was
'I can see that the hospital would not want to reveal her salary, if in fact she has one, and that has privacy implications,' Poses said. 'But I don’t understand why the hospital would not be able to simply tell you whether or not she is employed there and in what capacity.'
Thus, the latest commentary in the New England Journal of Medicine by National Correspondent Dr Lisa Rosenbaum used special pleading to argue for an expensive medical device with dubious benefits but which no more dubious evidence suggests may cause cancer. The commentary also decried the device's critics as unreasonably concerned about "greed" and "corruption," even though the device is clearly expensive, there are at least creditable allegations that the device makers used political influence to promote it and concealed its risks, and ironically the commentary itself obfuscated whether Dr Rosenbaum has a major financial relationship to the hospital that is being sued in connection with use of that device there.
Thus, this latest New England Journal of Medicinearticle, like those by the same author in the same journal which we have discussed before (here and here), seems more like a rant in a political blog than a scholarly article in the US' and perhaps the world's most prestigious scholarly medical journal. What is going on at the NEJM? What has happened to its editorial standards? Why should it continue to inspire such trust?
ADDENDUM (22 March, 2016) - See also comments in the 1BoringOldMan blog.
1. Rosenbaum L. N-of-1 policymaking - tragedy, trade-offs, and the demise of morcellation. N Engl J Med 2016; 374: 986-990. Link here.
2. Chittawar PB, Franik S, Pouwer AW et al. Minimally invasive surgical techniques versus open myomectomy for uterine fibroids. Cochrane Library 2014. Link here.
3. Dyer O. US surgeon who campaigned against potentially dangerous device receives legal threat. Brit Med J 2014; 349: g5577.